There is a history of healthcare providers complaining about clinical decision support (CDS) systems. “They’re too exhausting,” the provider may say. “What if they make a big mistake?”
According to a study published by the Journal of the American College of Radiology, providers may not always love ordering imaging services through a CDS system, but that unhappiness does not lead to a significant amount of “leakage,” or out-of-network referrals.
Anand M. Prabhakar, MD, Harvard Medical School and Massachusetts General Hospital, and colleagues studied data from their own medical center’s CDS system from Jan. 1, 2011, to Dec. 31, 2013, looking at studies that received low appropriateness scores and were then cancelled by the ordering physician.
Overall, the authors found that leakage of CTs and MRIs was more than 31 percent; more than 9 percent leaked to affiliated hospitals, and more than 21 percent leaked to hospitals and imaging centers completely out of their network.
However, If this number seems high, Prabhakar et al. said they consider anything in range of 20 to 30 percent to be normal. In addition, they noted that the amount of leakage remained fairly consistent throughout over the course of the three years being examined, “suggesting that consistent factors may be contributing to leakage.”
“Although leakage of imaging studies was a significant occurrence among population health management patients, order-entry CDS systems did not seem to contribute significantly to this leakage,” the authors wrote. “Therefore, institutions can consider implementing order-entry CDS for both Medicare and non-Medicare populations, with less fear of driving substantial imaging revenue outside of their institution.”
Prabhakar and colleagues noted that just 4.5 percent of studies receiving a low decision support score from the CDS “leaked” outside the medical center’s network. This shows that, if anything, use of a CDS system may be leading to less leakage.
“A host of potential advantages of order-entry CDS systems could attract imaging business,” the authors wrote. “For example, some insurers have accepted order-entry CDS scores for imaging preauthorization. As a result, ordering providers have the choice between an order-entry CDS process that should take no longer than a few minutes, and alternatively, the potentially onerous task of obtaining phone preauthorization from a radiology benefits management company. The potential time savings for ordering providers and their staff could be substantial.”
According to the authors, another advantage of using CDS systems is that providers receive real-time feedback and can learn from that feedback, helping their understanding of when different services are and are not necessary.
The authors did note that their study had multiple limitations. For example, the study only focused on a single medical center, so the results may not properly represent other institutions. Also, the study looked for leakage within 60 days of when the order was originally entered into the CDS system, and choosing a longer period of time may have produced different results.